Corneal nerve abnormalities in early-stage diabetic retinopathy evaluated by corneal confocal microscopy

This study aims to investigate the correlation between the development of diabetic retinopathy (DR) and the changes in corneal sub-basal nerve plexus (SNP) and corneal dendritic cells (DCs).

58 patients with type 2 diabetes mellitus (T2DM) and 30 age- and sex-matched healthy participants underwent assessment of the corneal nerve. The DR group was divided into no diabetic retinopathy (NDR) and 29 eyes with mild to moderate non-proliferative diabetic retinopathy (NPDR). We collected the clinical data (gender, age, body mass index), laboratory indicators (fasting blood glucose, glycosylated hemoglobin), and confocal laser scanning microscopy images for further analysis. Spearman rank correlation or Pearson correlation analyses were used to evaluate the relationships.

Compared to the control group, all parameters of corneal nerves in the case group were significantly reduced (all P < 0.001). Similarly, the density of immature dendritic cells (DCs) in the NDR group (70.40 [49.10-87.50]) was significantly higher than that in patients with DR (22.80 [14.30-34.50]), while the density of mature DCs (27.80 [17.70-39.60]) was significantly lower than that in patients with DR (70.80 [47.20-90.40]) (all P < 0.001). The morphological score of DCs in the NPDR group (2.6 [2.4-2.8]) was significantly higher than that in the NDR group (1.80 [1.80-2.20]) (all P < 0.001). As diabetes progresses to the stage of diabetic retinopathy, there were significant negative correlations between inferior whorl length (IWL) (rho = -0.802, P < 0.001) and average nerve fiber length ( ANFL) (rho = -0.487, P < 0.001) with the occurrence of DR. Additionally, there were significant positive correlations between the density of mature DCs (rho = 0.704, P < 0.001) and corneal dendritic morphology (DCM)(rho = 0.791, P < 0.001) with the occurrence of DR.

Compared to patients with NDR, those with NPDR exhibited a reduction in corneal nerve fiber density (CNFD)and corneal nerve fiber length (CNFL), and the length of corneal nerves in the inferior whorl (IW) was also reduced and showed a significant negative correlation with the presence of fundus lesions. Patients with DR had an increased density and morphological score of mature DCs in the cornea. These findings suggest that we should focus on the changes in corneal nerves when screening and diagnosing DR, and treatment should be considered for better prognosis.